Hey, I hope it's not weird to hear from the author of a book you reviewed, but I just stumbled across this and thought it was a really thoughtful review.
I do agree that BCIs just aren't progressing that fast. The most unrealistic technology in Nexus is the nexus technology itself. I knew that to a large extent when I wrote the books, but was so fascinated by the potential of neurotech (even if, in reality, it will take much much much longer to develop) that I wanted to explore it in fiction.
I was off somewhat on AI as well, but in my mind the jury is still out on that. I'm not convinced we're really in a "line go up" world even now. But I do acknowledge that my confidence in that is weaker than it was just a few years ago. Even so, realistically, in the year 2040 when the book is set, there would be much more influence of AI technologies than we see in the books. If and when I write more fiction, that'll be something I explore.
Regardless, I'm glad you enjoyed the book, and thanks for a thoughtful review.
Wow, thank you for your comment! I really enjoyed the book and hope you do write more. I think there's a range of reasonable views on how fast AI progress will continue, but given the uncertainty I think the world needs more near-future science fiction to build out these assumptions/consequences.
I do think we're in quite a "line go up" world and we have a lot of room to still scale RL, which makes me think 2040 might be far more strange than most people are imagining today. And it seems like there are lots of unsolved safety issues like chain of thought faithfulness, reward hacking, etc. We might need more stories like yours to shine an emotional light on the consequences of a very different 2040 or even 2027.
Hey, I hope it's not weird to hear from the author of a book you reviewed, but I just stumbled across this and thought it was a really thoughtful review.
I do agree that BCIs just aren't progressing that fast. The most unrealistic technology in Nexus is the nexus technology itself. I knew that to a large extent when I wrote the books, but was so fascinated by the potential of neurotech (even if, in reality, it will take much much much longer to develop) that I wanted to explore it in fiction.
I was off somewhat on AI as well, but in my mind the jury is still out on that. I'm not convinced we're really in a "line go up" world even now. But I do acknowledge that my confidence in that is weaker than it was just a few years ago. Even so, realistically, in the year 2040 when the book is set, there would be much more influence of AI technologies than we see in the books. If and when I write more fiction, that'll be something I explore.
Regardless, I'm glad you enjoyed the book, and thanks for a thoughtful review.
Wow, thank you for your comment! I really enjoyed the book and hope you do write more. I think there's a range of reasonable views on how fast AI progress will continue, but given the uncertainty I think the world needs more near-future science fiction to build out these assumptions/consequences.
I do think we're in quite a "line go up" world and we have a lot of room to still scale RL, which makes me think 2040 might be far more strange than most people are imagining today. And it seems like there are lots of unsolved safety issues like chain of thought faithfulness, reward hacking, etc. We might need more stories like yours to shine an emotional light on the consequences of a very different 2040 or even 2027.
For example:
https://x.com/matt_beard_/status/1912633735381676480
https://80000hours.org/agi/guide/when-will-agi-arrive/
Safety papers like Anthropic's alignment faking or OpenAI's detecting chain misbehavior also come to mind.
Anyway, thanks again for your comment and your work :)